Risk factors and prediction of very short term versus short/intermediate term post-stroke mortality: A data mining approach

Predicting mortality in brain stroke patients using neural networks: outcomes analysis in a longitudinal study

In this study, Neural Networks (NN) modelling has emerged as a promising tool for predicting outcomes in patients with Brain Stroke (BS) by identifying key risk factors. In this longitudinal study, we enrolled 332 patients form Imam hospital in Ardabil, Iran, with mean age: 77.4 (SD 10.4) years, and 50.6% were male. Diagnosis of BS was confirmed using both computerized tomography scan and magnetic resonance imaging, and risk factor and outcome data were collected from the hospital's BS registry, and by telephone follow-up over a period of 10 years, respectively. Using a multilayer perceptron NN approach, we analysed the impact of various risk factors on time to mortality and mortality from BS. A total of 100 NN classification algorithm were trained utilizing STATISTICA 13 software, and the optimal model was selected for further analysis based on their diagnostic performance. We also calculated Kaplan-Meier survival probabilities and conducted Log-rank tests. The five selected NN models exhibited impressive accuracy ranges of 81-85%. However, the optimal model stood out for its superior diagnostic indices. Mortality rate in the training and the validation data set was 7.9 (95% CI 5.7-11.0) per 1000 and 8.2 (7.1-9.6) per 1000, respectively (P = 0.925). The optimal model highlighted significant risk factors for BS mortality, including smoking, lower education, advanced age, lack of physical activity, a history of diabetes, all carrying substantial importance weights. Our study provides compelling evidence that the NN approach is highly effective in predicting mortality in patients with BS based on key risk factors, and has the potential to significantly enhance the accuracy of prediction. Moreover, our findings could inform more effective prevention strategies for BS, ultimately leading to better patient outcomes.

Stroke mortality prediction using machine learning: systematic review

BACKGROUND AND AIMS

Accurate prognostication of stroke may help in appropriate therapy and rehabilitation planning. In the past few years, several machine learning (ML) algorithms were applied for prediction of stroke outcomes. We aimed to examine the performance of machine learning-based models for the prediction of mortality after stroke, as well as to identify the most prominent factors for mortality.

MATERIALS AND METHODS

We searched MEDLINE/PubMed and Web of Science databases for original publications on machine learning applications in stroke mortality prediction, published between January 1, 2011, and October 27, 2022. Risk of bias and applicability were evaluated using the tailored QUADAS-2 tool.

RESULTS

Of the 1015 studies retrieved, 28 studies were included. Twenty-Five studies were retrospective. The ML models demonstrated a favorable range of AUC for mortality prediction (0.67-0.98). In most of the articles, the models were applied for short-term post stroke mortality. The number of explanatory features used in the models to predict mortality ranged from 5 to 200, with substantial overlap in the variables included. Age, high BMI and high NIHSS score were identified as important predictors for mortality. Almost all studies had a high risk of bias in at least one category and concerns regarding applicability.

CONCLUSION

Using machine learning, data available at the time of admission may aid in stroke mortality prediction. Notwithstanding, current research is based on few preliminary works with high risk of bias and high heterogeneity. Thus, future prospective, multicenter studies with standardized reports are crucial to firmly establish the usefulness of the algorithms in stroke prognostication.

Stacking ensemble learning model to predict 6-month mortality in ischemic stroke patients

Patients with acute ischemic stroke can benefit from reperfusion therapy. Nevertheless, there are gray areas where initiation of reperfusion therapy is neither supported nor contraindicated by the current practice guidelines. In these situations, a prediction model for mortality can be beneficial in decision-making. This study aimed to develop a mortality prediction model for acute ischemic stroke patients not receiving reperfusion therapies using a stacking ensemble learning model. The model used an artificial neural network as an ensemble classifier. Seven base classifiers were K-nearest neighbors, support vector machine, extreme gradient boosting, random forest, naive Bayes, artificial neural network, and logistic regression algorithms. From the clinical data in the International Stroke Trial database, we selected a concise set of variables assessable at the presentation. The primary study outcome was all-cause mortality at 6 months. Our stacking ensemble model predicted 6-month mortality with acceptable performance in ischemic stroke patients not receiving reperfusion therapy. The area under the curve of receiver-operating characteristics, accuracy, sensitivity, and specificity of the stacking ensemble classifier on a put-aside validation set were 0.783 (95% confidence interval 0.758-0.808), 71.6% (69.3-74.2), 72.3% (69.2-76.4%), and 70.9% (68.9-74.3%), respectively.

Familial Mortality Risks in Patients With Ischemic Stroke: A Swedish Sibling Study

BACKGROUND

The influence of familial factors on the prognosis of ischemic stroke (IS) is unknown. This nationwide follow-up study evaluated familial mortality risks of IS among Swedish sibling pairs hospitalized for IS.

METHODS

We linked Swedish nationwide registers for the identification of 1380 Swedish born sibling pairs (2760 cases), where both siblings were hospitalized for first-time IS between 1991 and 2010. Median age was 62 years (range, 26-78 years). Sibling pairs with cancer were excluded. Familial hazard ratios (FHRs) for mortality after first IS hospitalization were determined with Cox regression. The influence of proband survival after IS was categorized as short sibling survival (<1, 2, 3, 4, or 5 years) or long sibling survival (≥1, 2, 3, 4, or 5 years) after IS. FHRs were adjusted for age at onset, sex, education, county, year of diagnosis, days of hospitalization, and comorbidities.

RESULTS

Short sibling survival (ie, <3 or <4 years) after IS was associated with an adjusted FHR of 1.29 (95% CI, 1.05-1.58) and 1.24 (95% CI, 1.02-1.51), respectively, for overall mortality after IS. Stratified analysis showed that short sibling survival (ie, <2-<5 years) was stronger and significant only among younger individuals (<62 years) and males. Highest FHR for short sibling survival (ie, <4 years) was 1.42 (95% CI, 1.08-1.88) for younger individuals and 1.50 (95% CI, 1.21-1.87) for males. For young male subjects, FHR was 1.80 (95% CI, 1.33-2.46). In the adjusted model, mortality was also associated with sex, education, age at onset, year of diagnosis, days of hospitalization, coronary heart disease, diabetes, dementia, heart failure, obesity, alcoholism, and hyperlipidemia.

CONCLUSIONS

Our results suggest that family history of short survival in siblings after IS is associated with mortality after IS for younger male subjects. Additional studies are needed to characterize possible genetic and nongenetic familial environmental causes of this association.

A survey on data mining techniques used in medicine

Data mining is the process of analyzing a massive amount of data to identify meaningful patterns and detect relations, which can lead to future trend prediction and appropriate decision making. Data mining applications are significant in marketing, banking, medicine, etc. In this paper, we present an overview of data mining applications in medicine to provide a clear view of the challenges and previous works in this area for researchers. Data mining techniques such as Decision Tree, Random Forest, K-means Clustering, Support Vector Machine, Logistic Regression, Neural Network, Naive Bayes, and association rule mining are used for diagnosing, prognosis, classifying, constructing predictive models, and analyzing risk factors of various diseases. The main objective of the paper is to analyze and compare different data mining techniques used in the medical applications. We present a summary of the results and provide comparison analysis of the data mining methods employed by the reviewed articles.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40200-021-00884-2.

Predicting short and long-term mortality after acute ischemic stroke using EHR

OBJECTIVE

Despite improvements in treatment, stroke remains a leading cause of mortality and long-term disability. In this study, we leveraged administrative data to build predictive models of short- and long-term post-stroke all-cause-mortality.

METHODS

The study was conducted and reported according to the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) guideline. We used patient-level data from electronic health records, three algorithms, and six prediction windows to develop models for post-stroke mortality.

RESULTS

We included 7144 patients from which 5347 had survived their ischemic stroke after two years. The proportion of mortality was between 8%(605/7144) within 1-month, to 25%(1797/7144) for the 2-years window. The three most common comorbidities were hypertension, dyslipidemia, and diabetes. The best Area Under the ROC curve(AUROC) was reached with the Random Forest model at 0.82 for the 1-month prediction window. The negative predictive value (NPV) was highest for the shorter prediction windows - 0.91 for the 1-month - and the best positive predictive value (PPV) was reached for the 6-months prediction window at 0.92. Age, hemoglobin levels, and body mass index were the top associated factors. Laboratory variables had higher importance when compared to past medical history and comorbidities. Hypercoagulation state, smoking, and end-stage renal disease were more strongly associated with long-term mortality.

CONCLUSION

All the selected algorithms could be trained to predict the short and long-term mortality after stroke. The factors associated with mortality differed depending on the prediction window. Our classifier highlighted the importance of controlling risk factors, as indicated by laboratory measures.

Metabolic Physiological Networks: The Impact of Age

Metabolic homeostasis emerges from the interplay between several feedback systems that regulate the physiological variables related to energy expenditure and energy availability, maintaining them within a certain range. Although it is well known how each individual physiological system functions, there is little research focused on how the integration and adjustment of multiple systems results in the generation of metabolic health. The aim here was to generate an integrative model of metabolism, seen as a physiological network, and study how it changes across the human lifespan. We used data from a transverse, community-based study of an ethnically and educationally diverse sample of 2572 adults. Each participant answered an extensive questionnaire and underwent anthropometric measurements (height, weight, and waist), fasting blood tests (glucose, HbA1c, basal insulin, cholesterol HDL, LDL, triglycerides, uric acid, urea, and creatinine), along with vital signs (axillar temperature, systolic, and diastolic blood pressure). The sample was divided into 6 groups of increasing age, beginning with less than 25 years and increasing by decades up to more than 65 years. In order to model metabolic homeostasis as a network, we used these 15 physiological variables as nodes and modeled the links between them, either as a continuous association of those variables, or as a dichotomic association of their corresponding pathological states. Weight and overweight emerged as the most influential nodes in both types of networks, while high betweenness parameters, such as triglycerides, uric acid and insulin, were shown to act as gatekeepers between the affected physiological systems. As age increases, the loss of metabolic homeostasis is revealed by changes in the network’s topology that reflect changes in the system−wide interactions that, in turn, expose underlying health stages. Hence, specific structural properties of the network, such as weighted transitivity, i.e., the density of triangles in the network, can provide topological indicators of health that assess the whole state of the system. Overall, our findings show the importance of visualizing health as a network of organs and/or systems, and highlight the importance of triglycerides, insulin, uric acid and glucose as key biomarkers in the prevention of the development of metabolic disorders.

Geographical variations in maternal lifestyles during pregnancy associated with congenital heart defects among live births in Shaanxi province, Northwestern China

In this study, we aimed to explore regional differences in maternal lifestyle during pregnancy related to congenital heart defects (CHD) in Shaanxi province, Northwestern China. A large-scale epidemiologic survey of birth defects among infants born during 2010–2013, was conducted in Shaanxi province. Non-spatial and geographic weighted logistic regression models were used for analysis. The spatial model indicated that passive smoking frequency was positively associated with CHD for 43.3% of participants (P < 0.05), with the highest OR in North Shaanxi and the lowest in South Shaanxi. Approximately 49.2% of all mothers who ever drink tea were more likely to have an infant with CHD (P < 0.05), with the highest OR values observed in North and Central Shaanxi. Additionally, maternal alcohol intake frequency ≥ 1/week was significantly correlated with CHD among about 24.7% of participants (P < 0.05), with OR values ranging from 0.738 (Central Shaanxi) to 1.198 (North Shaanxi). The rates of unhealthy maternal lifestyles during pregnancy associated with CHD differed in various areas of the province. The role of geographical variations in these factors may provide some possible clues and basis for tailoring site-specific intervention strategies.

A systematic review of machine learning models for predicting outcomes of stroke with structured data

Background and purpose

Machine learning (ML) has attracted much attention with the hope that it could make use of large, routinely collected datasets and deliver accurate personalised prognosis. The aim of this systematic review is to identify and critically appraise the reporting and developing of ML models for predicting outcomes after stroke.

Methods

We searched PubMed and Web of Science from 1990 to March 2019, using previously published search filters for stroke, ML, and prediction models. We focused on structured clinical data, excluding image and text analysis. This review was registered with PROSPERO (CRD42019127154).

Results

Eighteen studies were eligible for inclusion. Most studies reported less than half of the terms in the reporting quality checklist. The most frequently predicted stroke outcomes were mortality (7 studies) and functional outcome (5 studies). The most commonly used ML methods were random forests (9 studies), support vector machines (8 studies), decision trees (6 studies), and neural networks (6 studies). The median sample size was 475 (range 70–3184), with a median of 22 predictors (range 4–152) considered. All studies evaluated discrimination with thirteen using area under the ROC curve whilst calibration was assessed in three. Two studies performed external validation. None described the final model sufficiently well to reproduce it.

Conclusions

The use of ML for predicting stroke outcomes is increasing. However, few met basic reporting standards for clinical prediction tools and none made their models available in a way which could be used or evaluated. Major improvements in ML study conduct and reporting are needed before it can meaningfully be considered for practice.

The Impact of Education and Age on Metabolic Disorders

Metabolic disorders, such as obesity, elevated blood pressure, dyslipidemias, insulin resistance, hyperglycemia, and hyperuricemia have all been identified as risk factors for an epidemic of important and widespread chronic-degenerative diseases, such as type 2 diabetes and cardiovascular disease, that constitute some of the world's most important public health challenges. Their increasing prevalence can be associated with an aging population and to lifestyles within an obesogenic environment. Taking educational level as a proxy for lifestyle, and using both logistic and linear regressions, we study the relation between a wide set of metabolic biomarkers, and educational level, body mass index (BMI), age, and sex as correlates, in a population of 1,073 students, academic and non-academic staff at Mexico's largest university (UNAM). Controlling for BMI and sex, we consider educational level and age as complementary measures-degree and duration-of exposure to metabolic insults. Analyzing the role of education across a wide spectrum of educational levels (from primary school to doctoral degree), we show that higher education correlates to significantly better metabolic health when compared to lower levels, and is associated with significantly less risk for waist circumference, systolic blood pressure, glucose, glycosylated hemoglobin, triglycerides, high density lipoprotein and metabolic syndrome (all p < 0.05); but not for diastolic blood pressure, basal insulin, uric acid, low density lipoprotein, and total cholesterol. We classify each biomarker, and corresponding metabolic disorder, by its associated set of statistically significant correlates. Differences among the sets of significant correlates indicate various aetiologies and the need for targeted population-specific interventions. Thus, variables strongly linked to educational level are candidates for lifestyle change interventions. Hence, public policy efforts should be focused on those metabolic biomarkers strongly linked to education, while adopting a different approach for those biomarkers not linked as they may be poor targets for educational campaigns.

Big data analytics for preventive medicine

Medical data is one of the most rewarding and yet most complicated data to analyze. How can healthcare providers use modern data analytics tools and technologies to analyze and create value from complex data? Data analytics, with its promise to efficiently discover valuable pattern by analyzing large amount of unstructured, heterogeneous, non-standard and incomplete healthcare data. It does not only forecast but also helps in decision making and is increasingly noticed as breakthrough in ongoing advancement with the goal is to improve the quality of patient care and reduces the healthcare cost. The aim of this study is to provide a comprehensive and structured overview of extensive research on the advancement of data analytics methods for disease prevention. This review first introduces disease prevention and its challenges followed by traditional prevention methodologies. We summarize state-of-the-art data analytics algorithms used for classification of disease, clustering (unusually high incidence of a particular disease), anomalies detection (detection of disease) and association as well as their respective advantages, drawbacks and guidelines for selection of specific model followed by discussion on recent development and successful application of disease prevention methods. The article concludes with open research challenges and recommendations.

Anthropometric measurements and mortality in frail older adults

BACKGROUND

As the number of older adults increases, so does the number of frail older adults. Although anthropometry has been widely used as a way to stratify the overall mortality risk of a person, the significance of these measurements becomes blurred in the case of frail older adults who have changes in body composition. Therefore, the aim of this study is to determine the association of anthropometric measurements (body mass index, knee-adjusted height body mass index, waist-to-hip ratio and calf circumference) with mortality risk in a group of older Mexican adults.

METHODS

This is a longitudinal analysis of the Mexican Health and Aging sub-sample (with biomarkers, n = 2573) from the first wave in 2001, followed-up to the last available wave in 2015. Only frail 50-year or older adults (Frailty Index with a cut-off value of 0.21 or higher, was used) were considered for this analysis (n = 1298). A survival analysis was performed with Kaplan-Meier curves and Cox regression models (unadjusted and adjusted for confounding). Socio-demographic, health risks, physical activity and comorbidities were variables used for adjusting the multivariate models.

RESULTS

From the total sample of 1298 older adults, 32.5% (n = 422) died during follow-up. The highest hazard ratio in the adjusted model was for calf circumference 1.31 (95% confidence interval 1.02-1.69, p = 0.034). Other measurements were not significant.

CONCLUSIONS

Anthropometric measurements have different significance in frail older adults, and these differences could have implications on adverse outcomes. Calf circumference has a potential value in predicting negative health outcomes.

Predicting post-stroke activities of daily living through a machine learning-based approach on initiating rehabilitation

OBJECTIVES

Prediction of activities of daily living (ADL) is crucial for optimized care of post-stroke patients. However, no suitably-validated and practical models are currently available in clinical practice.

METHODS

Participants of a Post-acute Care-Cerebrovascular Diseases (PAC-CVD) program from a reference hospital in Taiwan between 2014 and 2016 were enrolled in this study. Based on 15 rehabilitation assessments, machine learning (ML) methods, namely logistic regression (LR), support vector machine (SVM), and random forest (RF), were used to predict the Barthel index (BI) status at discharge. Furthermore, SVM and linear regression were used to predict the actual BI scores at discharge.

RESULTS

A total of 313 individuals (men: 208; women: 105) were enrolled in the study. All the classification models outperformed single assessments in predicting the BI statuses of the patients at discharge. The performance of the LR and RF algorithms was higher (area under ROC curve (AUC): 0.79) than that of SVM algorithm (AUC: 0.77). In addition, the mean absolute errors of both SVM and linear regression models in predicting the actual BI score at discharge were 9.86 and 9.95, respectively.

CONCLUSIONS

The proposed ML-based method provides a promising and practical computer-assisted decision making tool for predicting ADL in clinical practice.

Point-of-Care-Testing in Acute Stroke Management: An Unmet Need Ripe for Technological Harvest

Stroke, the second highest leading cause of death, is caused by an abrupt interruption of blood to the brain. Supply of blood needs to be promptly restored to salvage brain tissues from irreversible neuronal death. Existing assessment of stroke patients is based largely on detailed clinical evaluation that is complemented by neuroimaging methods. However, emerging data point to the potential use of blood-derived biomarkers in aiding clinical decision-making especially in the diagnosis of ischemic stroke, triaging patients for acute reperfusion therapies, and in informing stroke mechanisms and prognosis. The demand for newer techniques to deliver individualized information on-site for incorporation into a time-sensitive work-flow has become greater. In this review, we examine the roles of a portable and easy to use point-of-care-test (POCT) in shortening the time-to-treatment, classifying stroke subtypes and improving patient's outcome. We first examine the conventional stroke management workflow, then highlight situations where a bedside biomarker assessment might aid clinical decision-making. A novel stroke POCT approach is presented, which combines the use of quantitative and multiplex POCT platforms for the detection of specific stroke biomarkers, as well as data-mining tools to drive analytical processes. Further work is needed in the development of POCTs to fulfill an unmet need in acute stroke management.

Predicting congenital heart defects: A comparison of three data mining methods

Congenital heart defects (CHD) is one of the most common birth defects in China. Many studies have examined risk factors for CHD, but their predictive abilities have not been evaluated. In particular, few studies have attempted to predict risks of CHD from, necessarily unbalanced, population-based cross-sectional data. Therefore, we developed and validated machine learning models for predicting, before and during pregnancy, women’s risks of bearing children with CHD. We compared the results of these models in a large-scale, comprehensive population-based retrospective cross-sectional epidemiological survey of birth defects in six counties in Shanxi Province, China, covering 2006 to 2008. This contained 78 cases of CHD among 33831 live births. We constructed nine synthetic variables to use in the models: maternal age, annual per capita income, family history, maternal history of illness, nutrition and folic acid deficiency, maternal illness in pregnancy, medication use in pregnancy, environmental risk factors in pregnancy, and unhealthy maternal lifestyle in pregnancy. The machine learning algorithms Weighted Support Vector Machine (WSVM) and Weighted Random Forest (WRF) were trained on, and a logistic regression (Logit) was fitted to, two-thirds of the data. Their predictive abilities were then tested in the remaining data. True positive rate (TPR), true negative rate (TNR), accuracy (ACC), area under the curves (AUC), G-means, and Weighted accuracy (WTacc) were used to compare the classification performance of the models. Median values, from repeating the data partitioning 1000 times, were used in all comparisons. The TPR and TNR of the three classifiers were above 0.65 and 0.93, respectively, better than any reported in the literature. TPR, wtACC, AUC and G were highest for WSVM, showing that it performed best. All three models are precise enough to identify groups at high risk of CHD. They should all be considered for future investigations of other birth defects and diseases.

Age-series based link prediction in evolving disease networks

Recently, several research efforts based on social network analysis and methods have been made for medical care information. One of these efforts is to extract the relationships between diseases by using social network modeling. However, all of previous works used the relationships in a simple way in a network consisting of diseases regardless of time or age factors. In this paper, we predict the onset of future diseases on the basis of the current health status of patients by considering age factor. The problem of predicting the relations between diseases is a really difficult and, at the same time, an important task. For this purpose, this paper first constructs a weighted disease network and then, it proposes a novel link prediction method, to identify the connections between diseases, building the evolving structure of the disease network with respect to patients' ages. To the best of our knowledge, this is the first attempt in predicting the connections between diseases according to patients' ages. Experiments on a real network demonstrate that the proposed approach can reveal disease correlations accurately and perform well at capturing future disease risks.